Patients with metastatic pancreatic cancer (PC) survive 4-11 months following diagnosis, therefore, a new strategy for therapy for these patients is urgently needed. The innovative discovery of this revised competitive renewal proposal is that the insulin transcription factor pancreatic duodenal homeobox transcription factor (PDX-1) regulates proliferation and invasion of human PC cells, which assigns a new and important role for PDX-1. With this new role, PDX-1 could serve as a molecular target for PC therapy. We propose in this revised competitive renewal grant to 1) expand the definition of the PDX-1 profile to include both PDX-1 expression levels and PDX-1 genetic variation, 2) to analyze the PDX-1 profile in a much larger number of PC specimens, including an independent cohort, and 3) to relate the PDX-1 profile to how it might guide choice of multiple cycles of four PDX-1-based therapies in PC mouse models. Preliminary data also demonstrate that viral thymidine kinase expression using rat insulin promoter-thymidine kinase gene therapy (RIP-TK) in PC cells in mice can be detected in PC cells using PET and optical imaging with stable dual-labeled analogs of 9-(4-(18F) fluoro-3-hydroxymethylbutyl) guanine (18F-FHBG), therefore, response to therapy could be imaged. Hypothesis: Our hypothesis is that the PDX-1 profile in PC guides the choice of PDX-1-based therapies. Specific aim I: to determine whether a) the prognostic capability of expression levels of PDX- 1 in 2 cohorts of PC can be validated using QPCR, western blots and quantum dot deconvolution imaging and image analysis, b) PDX-1 genetic variation exists in PC using DNA sequencing and c) regulation of proliferation and invasion of human PC cell lines is influenced by PDX-1 expression levels and/or PDX-1 genetic variation. Specific aim II: to determine whether a) PDX-1 based therapies using 1) iv PDX-1 shRNA and/or 2) rat insulin promoter fragment-thymidine kinase and ganciclovir gene therapy have an effective cytoablative effect on PC in PC mouse models using viral and nonviral delivery systems, b) PDX-1 profile is altered after each cycle and thus guides choice of therapy and c) therapy-induced pancreatogenic (type 3) diabetes in mice is temporal with ultimate restoration of the pre-treatment PDX-1 profile, islet hormone levels and islet morphology and that the diabetes can be treated with insulin and/or pancreatic polypeptide administration. Specific aim III: to determine whether a) imaging agents to detect vTK expression using RIP-TK in PC cell lines with varying PDX-1 expression levels can be validated using 18F-FHBG for microPET imaging and/or dual labeled analogues for optical imaging and tomography and b) the imaged response to PDX-1-based therapies is dependent upon the PDX-1 profile. PUBLIC HEALTH RELEVANCE: Patients with metastatic pancreatic cancer (PC) survive 4-11 months following diagnosis, therefore, a new strategy for therapy for these patients is urgently needed. The innovative discovery of this proposal is that the insulin transcription factor pancreatic duodenal homeobox transcription factor (PDX-1) regulates human PC cells, which assigns a new and important role for PDX-1. Preliminary data support the hypothesis that PDX-1 is a therapeutic target for PC, which could be a meaningful advance for patients with this horrible disease.